From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 62AE2C43381 for ; Mon, 11 Mar 2019 15:19:43 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 2D0252084F for ; Mon, 11 Mar 2019 15:19:43 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727547AbfCKPTm (ORCPT ); Mon, 11 Mar 2019 11:19:42 -0400 Received: from mx2.suse.de ([195.135.220.15]:39644 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726352AbfCKPTl (ORCPT ); Mon, 11 Mar 2019 11:19:41 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.220.254]) by mx1.suse.de (Postfix) with ESMTP id 66452AFF3; Mon, 11 Mar 2019 15:19:39 +0000 (UTC) Subject: Re: [PATCH v4] btrfs-progs: dump-tree: add noscan option To: Anand Jain , linux-btrfs@vger.kernel.org References: <1551336865-9954-1-git-send-email-anand.jain@oracle.com> <6db79bc0-29a4-cdd1-73f3-9c433acb12f2@oracle.com> From: Nikolay Borisov Openpgp: preference=signencrypt Autocrypt: addr=nborisov@suse.com; prefer-encrypt=mutual; keydata= mQINBFiKBz4BEADNHZmqwhuN6EAzXj9SpPpH/nSSP8YgfwoOqwrP+JR4pIqRK0AWWeWCSwmZ T7g+RbfPFlmQp+EwFWOtABXlKC54zgSf+uulGwx5JAUFVUIRBmnHOYi/lUiE0yhpnb1KCA7f u/W+DkwGerXqhhe9TvQoGwgCKNfzFPZoM+gZrm+kWv03QLUCr210n4cwaCPJ0Nr9Z3c582xc bCUVbsjt7BN0CFa2BByulrx5xD9sDAYIqfLCcZetAqsTRGxM7LD0kh5WlKzOeAXj5r8DOrU2 GdZS33uKZI/kZJZVytSmZpswDsKhnGzRN1BANGP8sC+WD4eRXajOmNh2HL4P+meO1TlM3GLl EQd2shHFY0qjEo7wxKZI1RyZZ5AgJnSmehrPCyuIyVY210CbMaIKHUIsTqRgY5GaNME24w7h TyyVCy2qAM8fLJ4Vw5bycM/u5xfWm7gyTb9V1TkZ3o1MTrEsrcqFiRrBY94Rs0oQkZvunqia c+NprYSaOG1Cta14o94eMH271Kka/reEwSZkC7T+o9hZ4zi2CcLcY0DXj0qdId7vUKSJjEep c++s8ncFekh1MPhkOgNj8pk17OAESanmDwksmzh1j12lgA5lTFPrJeRNu6/isC2zyZhTwMWs k3LkcTa8ZXxh0RfWAqgx/ogKPk4ZxOXQEZetkEyTFghbRH2BIwARAQABtCNOaWtvbGF5IEJv cmlzb3YgPG5ib3Jpc292QHN1c2UuY29tPokCOAQTAQIAIgUCWIo48QIbAwYLCQgHAwIGFQgC CQoLBBYCAwECHgECF4AACgkQcb6CRuU/KFc0eg/9GLD3wTQz9iZHMFbjiqTCitD7B6dTLV1C ddZVlC8Hm/TophPts1bWZORAmYIihHHI1EIF19+bfIr46pvfTu0yFrJDLOADMDH+Ufzsfy2v HSqqWV/nOSWGXzh8bgg/ncLwrIdEwBQBN9SDS6aqsglagvwFD91UCg/TshLlRxD5BOnuzfzI Leyx2c6YmH7Oa1R4MX9Jo79SaKwdHt2yRN3SochVtxCyafDlZsE/efp21pMiaK1HoCOZTBp5 VzrIP85GATh18pN7YR9CuPxxN0V6IzT7IlhS4Jgj0NXh6vi1DlmKspr+FOevu4RVXqqcNTSS E2rycB2v6cttH21UUdu/0FtMBKh+rv8+yD49FxMYnTi1jwVzr208vDdRU2v7Ij/TxYt/v4O8 V+jNRKy5Fevca/1xroQBICXsNoFLr10X5IjmhAhqIH8Atpz/89ItS3+HWuE4BHB6RRLM0gy8 T7rN6ja+KegOGikp/VTwBlszhvfLhyoyjXI44Tf3oLSFM+8+qG3B7MNBHOt60CQlMkq0fGXd mm4xENl/SSeHsiomdveeq7cNGpHi6i6ntZK33XJLwvyf00PD7tip/GUj0Dic/ZUsoPSTF/mG EpuQiUZs8X2xjK/AS/l3wa4Kz2tlcOKSKpIpna7V1+CMNkNzaCOlbv7QwprAerKYywPCoOSC 7P25Ag0EWIoHPgEQAMiUqvRBZNvPvki34O/dcTodvLSyOmK/MMBDrzN8Cnk302XfnGlW/YAQ csMWISKKSpStc6tmD+2Y0z9WjyRqFr3EGfH1RXSv9Z1vmfPzU42jsdZn667UxrRcVQXUgoKg QYx055Q2FdUeaZSaivoIBD9WtJq/66UPXRRr4H/+Y5FaUZx+gWNGmBT6a0S/GQnHb9g3nonD jmDKGw+YO4P6aEMxyy3k9PstaoiyBXnzQASzdOi39BgWQuZfIQjN0aW+Dm8kOAfT5i/yk59h VV6v3NLHBjHVw9kHli3jwvsizIX9X2W8tb1SefaVxqvqO1132AO8V9CbE1DcVT8fzICvGi42 FoV/k0QOGwq+LmLf0t04Q0csEl+h69ZcqeBSQcIMm/Ir+NorfCr6HjrB6lW7giBkQl6hhomn l1mtDP6MTdbyYzEiBFcwQD4terc7S/8ELRRybWQHQp7sxQM/Lnuhs77MgY/e6c5AVWnMKd/z MKm4ru7A8+8gdHeydrRQSWDaVbfy3Hup0Ia76J9FaolnjB8YLUOJPdhI2vbvNCQ2ipxw3Y3c KhVIpGYqwdvFIiz0Fej7wnJICIrpJs/+XLQHyqcmERn3s/iWwBpeogrx2Lf8AGezqnv9woq7 OSoWlwXDJiUdaqPEB/HmGfqoRRN20jx+OOvuaBMPAPb+aKJyle8zABEBAAGJAh8EGAECAAkF AliKBz4CGwwACgkQcb6CRuU/KFdacg/+M3V3Ti9JYZEiIyVhqs+yHb6NMI1R0kkAmzsGQ1jU zSQUz9AVMR6T7v2fIETTT/f5Oout0+Hi9cY8uLpk8CWno9V9eR/B7Ifs2pAA8lh2nW43FFwp IDiSuDbH6oTLmiGCB206IvSuaQCp1fed8U6yuqGFcnf0ZpJm/sILG2ECdFK9RYnMIaeqlNQm iZicBY2lmlYFBEaMXHoy+K7nbOuizPWdUKoKHq+tmZ3iA+qL5s6Qlm4trH28/fPpFuOmgP8P K+7LpYLNSl1oQUr+WlqilPAuLcCo5Vdl7M7VFLMq4xxY/dY99aZx0ZJQYFx0w/6UkbDdFLzN upT7NIN68lZRucImffiWyN7CjH23X3Tni8bS9ubo7OON68NbPz1YIaYaHmnVQCjDyDXkQoKC R82Vf9mf5slj0Vlpf+/Wpsv/TH8X32ajva37oEQTkWNMsDxyw3aPSps6MaMafcN7k60y2Wk/ TCiLsRHFfMHFY6/lq/c0ZdOsGjgpIK0G0z6et9YU6MaPuKwNY4kBdjPNBwHreucrQVUdqRRm RcxmGC6ohvpqVGfhT48ZPZKZEWM+tZky0mO7bhZYxMXyVjBn4EoNTsXy1et9Y1dU3HVJ8fod 5UqrNrzIQFbdeM0/JqSLrtlTcXKJ7cYFa9ZM2AP7UIN9n1UWxq+OPY9YMOewVfYtL8M= Message-ID: <4c72cfcd-af49-5528-94b6-d7deaae72b03@suse.com> Date: Mon, 11 Mar 2019 17:19:38 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.4.0 MIME-Version: 1.0 In-Reply-To: <6db79bc0-29a4-cdd1-73f3-9c433acb12f2@oracle.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org On 11.03.19 г. 16:44 ч., Anand Jain wrote: > > > On 3/11/19 6:11 PM, Nikolay Borisov wrote: >> >> >> On 28.02.19 г. 8:54 ч., Anand Jain wrote: >>> From: Anand Jain >>> >>> There is no way to dump tree from a disk in the degraded mode. >>> Such as, when you specify a device for the cli 'btrfs inspect >>> dump-tree /dev/sda' it would invariably scan for its partner devices >>> to built and print the tree. This approach at times defeats the purpose. >> >> define "at times" which are those times? > >  Its already defined as below. Will reword or state it explicitly. > >>> Suppose in RAID1 if you want to review each mirror separately as of now >>> there is no way you can do that. So this patch adds an option --noscan >>> for example: >>>    'btrfs inspect dump-tree --noscan [..]' >>> with which it shall not scan for the other devices to report the >>> tree, and >>> helps debug RAID1s. >> >> This example contradicts what the code is doing, > >  No. did you miss the flag OPEN_CTREE_NO_DEVICES at [1] below. ? >  Which means it does not scan the system for the btrfs devices. > >> with this patch you >> will call btrfs_scan_one_device for every device passed to the noscan >> option? > >  That's right. Its correct behavior. > >>> >>> Signed-off-by: Anand Jain >>> v3->v4: change the patch title. >>>     collapse scan_args() to its only parent cmd_inspect_dump_tree() >>>     (it was bit confusing). >>>     update the change log. >>>     update usage. >>>     update man page. >>> v2->v3: make it scalable for more than two disks in noscan mode >>> v1->v2: rename --degraded to --noscan >> >> So what problem is this actually trying to solve? The following succeeds >> for me with btrfs-progs 4.17.1: > >> root@ubuntu-virtual:~# losetup -f file1.img >> root@ubuntu-virtual:~# losetup -f file2.img >> root@ubuntu-virtual:~# mkfs.btrfs -m raid1 /dev/loop1 /dev/loop0 >> losetup -d /dev/loop1 >> root@ubuntu-virtual:~# mount /dev/loop0 /media/scratch/ >> mount: wrong fs type, bad option, bad superblock on /dev/loop0, >>         missing codepage or helper program, or other error >> >>         In some cases useful info is found in syslog - try >>         dmesg | tail or so. >> >> >> root@ubuntu-virtual:~# btrfs inspect-internal dump-tree /dev/loop0 >> btrfs-progs v4.17.1 >> warning, device 1 is missing >> root tree >> leaf 30556160 items 10 free space 13173 generation 5 owner ROOT_TREE >> leaf 30556160 flags 0x1(WRITTEN) backref revision 1 >> fs uuid 8727408c-b1af-4c14-b63b-51b12bcff933 >> chunk uuid 5ebde928-324f-42d0-98c6-f953d680fc21 >>     item 0 key (EXTENT_TREE ROOT_ITEM 0) itemoff 15844 itemsize 439 >>         generation 5 root_dirid 0 bytenr 30523392 level 0 refs 1 >> ................. >> >> >> >> Your changelog definitely needs rewriting to make it clear what you are >> trying to do. > >  In your example above /dev/loop1 is destroyed. So are you suggesting to >  destroy the partner device to dump-tree in degraded mode? That's not >  a viable workaround. >From your changelog I was left with the impression it's not possible to issue dump-tree for a device whose counterpart is missing. Clearly that's not what you are fixing so you have to be more eloquent in your changelogs if you want people to quickly grasp what you are trying to do. > > Thanks, Anand > >>> --- >>>   Documentation/btrfs-inspect-internal.asciidoc |  5 ++- >>>   cmds-inspect-dump-tree.c                      | 57 >>> ++++++++++++++++++++------- >>>   2 files changed, 47 insertions(+), 15 deletions(-) >>> >>> diff --git a/Documentation/btrfs-inspect-internal.asciidoc >>> b/Documentation/btrfs-inspect-internal.asciidoc >>> index 381497d284b8..f9d7f1c58f00 100644 >>> --- a/Documentation/btrfs-inspect-internal.asciidoc >>> +++ b/Documentation/btrfs-inspect-internal.asciidoc >>> @@ -61,7 +61,7 @@ specify which mirror to print, valid values are 0, >>> 1 and 2 and the superblock >>>   must be present on the device with a valid signature, can be used >>> together with >>>   '--force' >>>   -*dump-tree* [options] :: >>> +*dump-tree* [options] [device...]:: >>>   (replaces the standalone tool *btrfs-debug-tree*) >>>   + >>>   Dump tree structures from a given device in textual form, expand >>> keys to human >>> @@ -95,6 +95,9 @@ intermixed in the output >>>   --bfs:::: >>>   use breadth-first search to print trees. the nodes are printed >>> before all >>>   leaves >>> +--noscan:::: >>> +do not scan the system for other partner device(s), only use the >>> device(s) >>> +provided in the argument >>>   -t :::: >>>   print only the tree with the specified ID, where the ID can be >>> numerical or >>>   common name in a flexible human readable form >>> diff --git a/cmds-inspect-dump-tree.c b/cmds-inspect-dump-tree.c >>> index ad5345b4f1db..9dd040e5de17 100644 >>> --- a/cmds-inspect-dump-tree.c >>> +++ b/cmds-inspect-dump-tree.c >>> @@ -21,6 +21,7 @@ >>>   #include >>>   #include >>>   #include >>> +#include >>>     #include "kerncompat.h" >>>   #include "radix-tree.h" >>> @@ -185,7 +186,7 @@ static u64 treeid_from_string(const char *str, >>> const char **end) >>>   } >>>     const char * const cmd_inspect_dump_tree_usage[] = { >>> -    "btrfs inspect-internal dump-tree [options] device", >>> +    "btrfs inspect-internal dump-tree [options] [ >>> ..]", >>>       "Dump tree structures from a given device", >>>       "Dump tree structures from a given device in textual form, >>> expand keys to human", >>>       "readable equivalents where possible.", >>> @@ -200,6 +201,7 @@ const char * const cmd_inspect_dump_tree_usage[] = { >>>       "-b|--block print info from the specified block only", >>>       "-t|--tree     print only tree with the given id >>> (string or number)", >>>       "--follow               use with -b, to show all children tree >>> blocks of ", >>> +    "--noscan               do not scan for the partner device(s)", >>>       NULL >>>   }; >>>   @@ -214,7 +216,7 @@ int cmd_inspect_dump_tree(int argc, char **argv) >>>       struct btrfs_disk_key disk_key; >>>       struct btrfs_key found_key; >>>       char uuidbuf[BTRFS_UUID_UNPARSED_SIZE]; >>> -    int ret; >>> +    int ret = 0; >>>       int slot; >>>       int extent_only = 0; >>>       int device_only = 0; >>> @@ -222,6 +224,7 @@ int cmd_inspect_dump_tree(int argc, char **argv) >>>       int roots_only = 0; >>>       int root_backups = 0; >>>       int traverse = BTRFS_PRINT_TREE_DEFAULT; >>> +    int dev_optind; >>>       unsigned open_ctree_flags; >>>       u64 block_only = 0; >>>       struct btrfs_root *tree_root_scan; >>> @@ -239,8 +242,8 @@ int cmd_inspect_dump_tree(int argc, char **argv) >>>       optind = 0; >>>       while (1) { >>>           int c; >>> -        enum { GETOPT_VAL_FOLLOW = 256, GETOPT_VAL_DFS, >>> -               GETOPT_VAL_BFS }; >>> +        enum { GETOPT_VAL_FOLLOW = 256, GETOPT_VAL_DFS, GETOPT_VAL_BFS, >>> +               GETOPT_VAL_NOSCAN}; >>>           static const struct option long_options[] = { >>>               { "extents", no_argument, NULL, 'e'}, >>>               { "device", no_argument, NULL, 'd'}, >>> @@ -252,6 +255,7 @@ int cmd_inspect_dump_tree(int argc, char **argv) >>>               { "follow", no_argument, NULL, GETOPT_VAL_FOLLOW }, >>>               { "bfs", no_argument, NULL, GETOPT_VAL_BFS }, >>>               { "dfs", no_argument, NULL, GETOPT_VAL_DFS }, >>> +            { "noscan", no_argument, NULL, GETOPT_VAL_NOSCAN }, >>>               { NULL, 0, NULL, 0 } >>>           }; >>>   @@ -313,24 +317,49 @@ int cmd_inspect_dump_tree(int argc, char **argv) >>>           case GETOPT_VAL_BFS: >>>               traverse = BTRFS_PRINT_TREE_BFS; >>>               break; >>> +        case GETOPT_VAL_NOSCAN: >>> +            open_ctree_flags |= OPEN_CTREE_NO_DEVICES;  <--- [1] >>> +            break; >>>           default: >>>               usage(cmd_inspect_dump_tree_usage); >>>           } >>>       } >>>   -    if (check_argc_exact(argc - optind, 1)) >>> +    if (check_argc_min(argc - optind, 1)) >>>           usage(cmd_inspect_dump_tree_usage); >>>   -    ret = check_arg_type(argv[optind]); >>> -    if (ret != BTRFS_ARG_BLKDEV && ret != BTRFS_ARG_REG) { >>> -        if (ret < 0) { >>> -            errno = -ret; >>> -            error("invalid argument %s: %m", argv[optind]); >>> -        } else { >>> -            error("not a block device or regular file: %s", >>> -                  argv[optind]); >>> +    dev_optind = optind; >>> +    while (dev_optind < argc) { >>> +        int fd; >>> +        struct btrfs_fs_devices *fs_devices; >>> +        u64 num_devices; >>> + >>> +        ret = check_arg_type(argv[optind]); >>> +        if (ret != BTRFS_ARG_BLKDEV && ret != BTRFS_ARG_REG) { >>> +            if (ret < 0) { >>> +                errno = -ret; >>> +                error("invalid argument %s: %m", argv[dev_optind]); >>> +            } else { >>> +                error("not a block device or regular file: %s", >>> +                       argv[dev_optind]); >>> +            } >>>           } >>> -        goto out; >>> +        fd = open(argv[dev_optind], O_RDONLY); >>> +        if (fd < 0) { >>> +            error("cannot open %s: %m", argv[dev_optind]); >>> +            return -EINVAL; >>> +        } >>> +        ret = btrfs_scan_one_device(fd, argv[dev_optind], &fs_devices, >>> +                        &num_devices, >>> +                        BTRFS_SUPER_INFO_OFFSET, >>> +                        SBREAD_DEFAULT); >>> +        close(fd); >>> +        if (ret) { >>> +            error("device scan %s: %s", argv[dev_optind], >>> +                   strerror(-ret)); >>> +            return ret; >>> +        } >>> +        dev_optind++; >>>       } >>>         printf("%s\n", PACKAGE_STRING); >>> >